The choice of a particular Artificial Neural Network (ANN) structure is a seemingly difficult task;worthy of relevance is that there is no systematic way for establishing a suitable architecture. In view of this, the ...The choice of a particular Artificial Neural Network (ANN) structure is a seemingly difficult task;worthy of relevance is that there is no systematic way for establishing a suitable architecture. In view of this, the study looked at the effects of ANN structural complexity and data pre-processing regime on its forecast performance. To address this aim, two ANN structural configurations: </span><b><span style="font-family:Verdana;font-size:12px;">1) Single-hidden layer, </span></b><span style="font-family:Verdana;font-size:12px;">and</span><b><span style="font-family:Verdana;font-size:12px;"> 2) Double-hidden layer</span></b><span style="font-family:Verdana;font-size:12px;"> feed</span></span><span style="font-family:Verdana;">-</span><span style="font-family:Verdana;">forward back</span><span style="font-size:10pt;font-family:""> </span><span style="font-size:10pt;font-family:""><span style="font-family:Verdana;font-size:12px;">propagation network were employed. Results obtained revealed generally that: a) ANN comprised of double hidden layers tends to be less robust and converges with less accuracy than its single-hidden layer counterpart under identical situations;b) for a univariate time series, phase-space reconstruction using embedding dimension which is based on dynamical systems theory is an effective way for determining the appropriate number of ANN input neurons, and c) data pre-processing via the scaling approach excessively limits the output range of the transfer function. In specific terms considering extreme flow prediction capability on the basis of effective correlation: Percent maximum and minimum correlation coefficient (</span><b><span style="font-family:Verdana;font-size:12px;">R</span><sub><span style="font-family:Verdana;font-size:12px;">max</span></sub><span style="font-family:Verdana;font-size:12px;">%</span></b><span style="font-family:Verdana;font-size:12px;"> and </span><b><span style="font-family:Verdana;font-size:12px;">R</span><sub><span style="font-family:Verdana;font-size:12px;">min</span></sub><span style="font-family:Verdana;font-size:12px;">%</span></b><span style="font-family:Verdana;font-size:12px;">), on the average for one-day ahead forecast during the training and validation phases respectively for the adopted network structures: </span><b><span style="font-family:Verdana;font-size:12px;">8 7 5 (</span><i><span style="font-family:Verdana;font-size:12px;">i.e.</span></i><span style="font-family:Verdana;font-size:12px;">, 8 input nodes, 7 nodes in the hidden layer, and 5 output nodes in the output layer)</span></b><span style="font-family:Verdana;font-size:12px;">, </span><b><span style="font-family:Verdana;font-size:12px;">8 5 2 5 (8 nodes in the input layer, 5 nodes in the first hidden layer, 2 nodes in the second hidden layer, and 5 nodes in the output layer)</span></b><span style="font-family:Verdana;font-size:12px;">, and </span><b><span style="font-family:Verdana;font-size:12px;">8 4 3 5 (8 nodes in the input layer, 4 nodes in the first hidden layer, 3 nodes in the second hidden layer, and 5 nodes in the output layer)</span></b><span style="font-family:Verdana;font-size:12px;"> gave: </span><b><span style="font-family:Verdana;font-size:12px;">101.2</span></b><span style="font-family:Verdana;font-size:12px;">, </span><b><span style="font-family:Verdana;font-size:12px;">99.4</span></b><span style="font-family:Verdana;font-size:12px;">;</span><b><span style="font-family:Verdana;font-size:12px;">100.2</span></b><span style="font-family:Verdana;font-size:12px;">, </span><b><span style="font-family:Verdana;font-size:12px;">218.3</span></b><span style="font-family:Verdana;font-size:12px;">;</span><b><span style="font-family:Verdana;font-size:12px;">93.7</span></b><span style="font-family:Verdana;font-size:12px;">, </span><b><span style="font-family:Verdana;font-size:12px;">95.0</span></b><span style="font-family:Verdana;font-size:12px;"> in all instances irrespective of the training algorithm (</span><i><span style="font-family:Verdana;font-size:12px;">i.e.</span></i><span style="font-family:Verdana;font-size:12px;">, pooled). On the other hand, in terms of percent of correct event prediction, the respective performances of the models for both low and high flows during the training and validation phases, respectively were: </span><b><span style="font-family:Verdana;font-size:12px;">0.78, 0.96: 0.65, 0.87;0.76, 0.93: 0.61, 0.83;</span></b><span style="font-family:Verdana;font-size:12px;">and</span><b><span style="font-family:Verdana;font-size:12px;"> 0.79, 0.96: 0.65, 0.87</span></b><span style="font-family:Verdana;font-size:12px;">. Thus, it suffices to note that on the basis of coherence or regularity of prediction consistency, the ANN model: </span><b><span style="font-family:Verdana;font-size:12px;">8 4 3 5</span></b><span style="font-family:Verdana;font-size:12px;"> performed better. This implies that though the adoption of large hidden layers vis-à-vis corresponding large neuronal signatures could be counter-productive because of network over-fitting, however, it may provide additional representational power. Based on the findings, it is imperative to note that ANN model is by no means a substitute for conceptual watershed </span></span><span style="font-family:Verdana;">modelling, therefore, exogenous variables should be incorporated in streamflow modelling and forecasting exercise because of their hydrologic evolutions.展开更多
To explore the variations in symbiotic N2 fixation and water use efficiency in cowpea, this study evaluated 25 USDA cowpea genotypes subjected to drought under field conditions at two locations (Kpachi and Woribogu) i...To explore the variations in symbiotic N2 fixation and water use efficiency in cowpea, this study evaluated 25 USDA cowpea genotypes subjected to drought under field conditions at two locations (Kpachi and Woribogu) in the Northern region of Ghana. The 15N and 13C natural abundance techniques were respectively used to assess N2 fixation and water use efficiency. The test genotypes elicited high symbiotic dependence in association with indigenous rhizobia, deriving between 55% and 98% of their N requirements from symbiosis. Consequently, the amounts of N-fixed by the genotypes showed remarkable variations, with values ranging from 37 kg·N-fixed·ha-1 to 337 kg·N-fixed·ha-1. Most genotypes elicited contrasting symbiotic performance between locations, a finding that highlights the effect of complex host/soil microbiome compatibility on the efficiency of the cowpea-rhizobia symbiosis. The test genotypes showed marked variations in water use efficiency, with most of the genotypes recording higher δ13C values when planted at Kpachi. Despite the high symbiotic dependence, the grain yield of the test cowpeas was low due to the imposed drought, and ranged from 56 kg/ha to 556 kg/ha at Kpachi, and 143 kg/ha to 748 kg/ha at Woribogu. The fact that some genotypes could grow and produce grain yields of 627 - 748 kg/ha under drought imposition is an important trait that could be tapped for further improvement of cowpea. These findings highlight the importance of the cowpea-rhizobia symbiosis and enhanced water relations in the crop’s wider adaptation to adverse edaphoclimatic conditions.展开更多
The novel coronavirus disease 2019(COVID-19)is the third coronavirus outbreak in the last two decades.Emerging and re-emerging infections like COVID-19 pose serious challenges of the paucity of information and lack of...The novel coronavirus disease 2019(COVID-19)is the third coronavirus outbreak in the last two decades.Emerging and re-emerging infections like COVID-19 pose serious challenges of the paucity of information and lack of specific cure or vaccines.This leaves utilisation of existing scientific data on related viral infections and repurposing relevant aetiologic and supportive therapies as the best control approach while novel strategies are developed and trialled.Many promising antiviral agents including lopinavir,ritonavir,remdesivir,umifenovir,darunavir,and oseltamivir have been repurposed and are currently trialled for the care for COVID-19 patients.Adjunct therapies for the management of symptoms and to provide support especially in severe and critically ill patients have also been identified.This review provides an appraisal of the current evidence for the rational use of frontline therapeutics in the management of COVID-19.It also includes updates regarding COVID-19 immunotherapy and vaccine development.展开更多
文摘The choice of a particular Artificial Neural Network (ANN) structure is a seemingly difficult task;worthy of relevance is that there is no systematic way for establishing a suitable architecture. In view of this, the study looked at the effects of ANN structural complexity and data pre-processing regime on its forecast performance. To address this aim, two ANN structural configurations: </span><b><span style="font-family:Verdana;font-size:12px;">1) Single-hidden layer, </span></b><span style="font-family:Verdana;font-size:12px;">and</span><b><span style="font-family:Verdana;font-size:12px;"> 2) Double-hidden layer</span></b><span style="font-family:Verdana;font-size:12px;"> feed</span></span><span style="font-family:Verdana;">-</span><span style="font-family:Verdana;">forward back</span><span style="font-size:10pt;font-family:""> </span><span style="font-size:10pt;font-family:""><span style="font-family:Verdana;font-size:12px;">propagation network were employed. Results obtained revealed generally that: a) ANN comprised of double hidden layers tends to be less robust and converges with less accuracy than its single-hidden layer counterpart under identical situations;b) for a univariate time series, phase-space reconstruction using embedding dimension which is based on dynamical systems theory is an effective way for determining the appropriate number of ANN input neurons, and c) data pre-processing via the scaling approach excessively limits the output range of the transfer function. In specific terms considering extreme flow prediction capability on the basis of effective correlation: Percent maximum and minimum correlation coefficient (</span><b><span style="font-family:Verdana;font-size:12px;">R</span><sub><span style="font-family:Verdana;font-size:12px;">max</span></sub><span style="font-family:Verdana;font-size:12px;">%</span></b><span style="font-family:Verdana;font-size:12px;"> and </span><b><span style="font-family:Verdana;font-size:12px;">R</span><sub><span style="font-family:Verdana;font-size:12px;">min</span></sub><span style="font-family:Verdana;font-size:12px;">%</span></b><span style="font-family:Verdana;font-size:12px;">), on the average for one-day ahead forecast during the training and validation phases respectively for the adopted network structures: </span><b><span style="font-family:Verdana;font-size:12px;">8 7 5 (</span><i><span style="font-family:Verdana;font-size:12px;">i.e.</span></i><span style="font-family:Verdana;font-size:12px;">, 8 input nodes, 7 nodes in the hidden layer, and 5 output nodes in the output layer)</span></b><span style="font-family:Verdana;font-size:12px;">, </span><b><span style="font-family:Verdana;font-size:12px;">8 5 2 5 (8 nodes in the input layer, 5 nodes in the first hidden layer, 2 nodes in the second hidden layer, and 5 nodes in the output layer)</span></b><span style="font-family:Verdana;font-size:12px;">, and </span><b><span style="font-family:Verdana;font-size:12px;">8 4 3 5 (8 nodes in the input layer, 4 nodes in the first hidden layer, 3 nodes in the second hidden layer, and 5 nodes in the output layer)</span></b><span style="font-family:Verdana;font-size:12px;"> gave: </span><b><span style="font-family:Verdana;font-size:12px;">101.2</span></b><span style="font-family:Verdana;font-size:12px;">, </span><b><span style="font-family:Verdana;font-size:12px;">99.4</span></b><span style="font-family:Verdana;font-size:12px;">;</span><b><span style="font-family:Verdana;font-size:12px;">100.2</span></b><span style="font-family:Verdana;font-size:12px;">, </span><b><span style="font-family:Verdana;font-size:12px;">218.3</span></b><span style="font-family:Verdana;font-size:12px;">;</span><b><span style="font-family:Verdana;font-size:12px;">93.7</span></b><span style="font-family:Verdana;font-size:12px;">, </span><b><span style="font-family:Verdana;font-size:12px;">95.0</span></b><span style="font-family:Verdana;font-size:12px;"> in all instances irrespective of the training algorithm (</span><i><span style="font-family:Verdana;font-size:12px;">i.e.</span></i><span style="font-family:Verdana;font-size:12px;">, pooled). On the other hand, in terms of percent of correct event prediction, the respective performances of the models for both low and high flows during the training and validation phases, respectively were: </span><b><span style="font-family:Verdana;font-size:12px;">0.78, 0.96: 0.65, 0.87;0.76, 0.93: 0.61, 0.83;</span></b><span style="font-family:Verdana;font-size:12px;">and</span><b><span style="font-family:Verdana;font-size:12px;"> 0.79, 0.96: 0.65, 0.87</span></b><span style="font-family:Verdana;font-size:12px;">. Thus, it suffices to note that on the basis of coherence or regularity of prediction consistency, the ANN model: </span><b><span style="font-family:Verdana;font-size:12px;">8 4 3 5</span></b><span style="font-family:Verdana;font-size:12px;"> performed better. This implies that though the adoption of large hidden layers vis-à-vis corresponding large neuronal signatures could be counter-productive because of network over-fitting, however, it may provide additional representational power. Based on the findings, it is imperative to note that ANN model is by no means a substitute for conceptual watershed </span></span><span style="font-family:Verdana;">modelling, therefore, exogenous variables should be incorporated in streamflow modelling and forecasting exercise because of their hydrologic evolutions.
文摘To explore the variations in symbiotic N2 fixation and water use efficiency in cowpea, this study evaluated 25 USDA cowpea genotypes subjected to drought under field conditions at two locations (Kpachi and Woribogu) in the Northern region of Ghana. The 15N and 13C natural abundance techniques were respectively used to assess N2 fixation and water use efficiency. The test genotypes elicited high symbiotic dependence in association with indigenous rhizobia, deriving between 55% and 98% of their N requirements from symbiosis. Consequently, the amounts of N-fixed by the genotypes showed remarkable variations, with values ranging from 37 kg·N-fixed·ha-1 to 337 kg·N-fixed·ha-1. Most genotypes elicited contrasting symbiotic performance between locations, a finding that highlights the effect of complex host/soil microbiome compatibility on the efficiency of the cowpea-rhizobia symbiosis. The test genotypes showed marked variations in water use efficiency, with most of the genotypes recording higher δ13C values when planted at Kpachi. Despite the high symbiotic dependence, the grain yield of the test cowpeas was low due to the imposed drought, and ranged from 56 kg/ha to 556 kg/ha at Kpachi, and 143 kg/ha to 748 kg/ha at Woribogu. The fact that some genotypes could grow and produce grain yields of 627 - 748 kg/ha under drought imposition is an important trait that could be tapped for further improvement of cowpea. These findings highlight the importance of the cowpea-rhizobia symbiosis and enhanced water relations in the crop’s wider adaptation to adverse edaphoclimatic conditions.
文摘The novel coronavirus disease 2019(COVID-19)is the third coronavirus outbreak in the last two decades.Emerging and re-emerging infections like COVID-19 pose serious challenges of the paucity of information and lack of specific cure or vaccines.This leaves utilisation of existing scientific data on related viral infections and repurposing relevant aetiologic and supportive therapies as the best control approach while novel strategies are developed and trialled.Many promising antiviral agents including lopinavir,ritonavir,remdesivir,umifenovir,darunavir,and oseltamivir have been repurposed and are currently trialled for the care for COVID-19 patients.Adjunct therapies for the management of symptoms and to provide support especially in severe and critically ill patients have also been identified.This review provides an appraisal of the current evidence for the rational use of frontline therapeutics in the management of COVID-19.It also includes updates regarding COVID-19 immunotherapy and vaccine development.